Search

Deep learning has shown tremendous prospects in human activity recognition based on micro-Doppler radar. However, the deficiency of radar data because of the extremely high labor and costs involved in obtaining radar data has hindered deep learning-based radar applications. In this paper, we proposed to use the discriminator of the generative adversarial network(GAN) for the radar-based human activities classification with low training data support. In supervised adversarial training, the generated samples and real samples are both sent to the discriminator for joint training, and thus the performance of the discriminator is boosted by both the real samples and generated samples. When the training samples are limited, GAN might fail to converge. In this paper, we firstly proposed a novel parameters initialization method of GAN to speed its convergence. The trained discriminator is fine-tuned on the training set for a few epochs. The classification results indicate that our proposed model can improve the classification accuracy and outperform the state-of-art models, especially when the training samples are limited.

Diabetic retinopathy (DR) is an important cause of blindness worldwide. However, DR is hard to be detected in the early stages, and the diagnostic procedure can be time-consuming even for the experienced experts. Therefore, a computer-aided diagnosis method based on deep learning algorithms is proposed to automatedly diagnose the referable diabetic retinopathy by classifying color retinal fundus photographs into two grades. In this paper, a novel convolutional neural network model with the Siamese-like architecture is trained with a transfer learning technique. Different from the previous works, the proposed model accepts binocular fundus images as inputs and learns their correlation to help to make a prediction. In the case with a training set of only 28 104 images and a test set of 7024 images, an area under the receiver operating curve of 0.951 is obtained by the proposed binocular model, which is 0.011 higher than that obtained by the existing monocular model. To further verify the effectiveness of the binocular design, a binocular model for five-class DR detection is also trained and evaluated on a 10% validation set. The result shows that it achieves a kappa score of 0.829 which is higher than that of the existing non-ensemble model.

In this paper, we present a novel one-dimensional deep convolutional neural network (1D-DCNN) with a multi-label-way-based algorithm for comprehensively and automatically extracting features and classifying mixture gases. Although a number of pattern recognition methods have been used to analyze the mixed gases, the performances of these methods highly depend on the hand-crafted feature engineering. By contrast, the proposed implementation, based on one-dimensional convolution, is capable of automatically extracting features and distinguishing the individual component of binary mixture gases composed of ethylene, CO, and methane. To the best of our knowledge, the proposed 1D-DCNN algorithm is first applied in the mixture gases’ recognition. In addition, the proposed 1D-DCNN with multi-label way not only significantly reduces the label dimension but also quantifies the probability of each component in mixed gases. Compared with the conventional pattern recognition algorithms including support vector machine, artificial neural network, k-nearest neighbor, and random forest, the proposed 1D-DCNN exhibits a higher recognition accuracy (96.30%) based on our extensive experimental results using ten-fold cross validation.

Diabetic retinopathy (DR) is a common complication of diabetes which is one of the leading causes of blindness worldwide. However, DR is hard to detect in the early stages and the diagnostic procedure can be time-consuming and abundant expertise is needed. Therefore, we proposed a computer-aided diagnosis method based on a deep learning algorithm to automatically diagnose DR and divide color retinal fundus photographs in to five grades. Besides, a novel pre-processing algorithm is adopted to enhance the quality and uniformity of input retinal images and a transfer learning method to achieve better performance. Finally, the system is evaluated based on a test set with 7023 images and an accuracy of 80.0% and a kappa score of 0.64 are achieved.

In this paper, we present four denoising algorithms dedicated to the division of focal plane (DoFP) polarization image sensors, including the average filtering, median filtering, Wiener filtering and wavelet threshold denoising algorithms. Compared with the previous implementations solely based on the Gaussian noise model, this paper, for the first time, covers the non-Gaussian noises, such as the salt & pepper noise and Poisson noise. According to our extensive experimental results, the wavelet threshold denoising outperforms for suppressing the Gaussian noise; while the median filtering and the Wiener filtering outperform for suppressing the low-density salt & pepper noise and Poisson noise, respectively.

In this paper, we present a novel one-dimensional deep convolutional neural network (1D-DCNN) based algorithm for classifying mixture gases. Compared with the previously reported electronic nose system that can only recognize pure gas, the proposed implementation is capable of distinguishing the individual component of binary mixture gases composed of Ethylene, CO and Methane. To the best of our knowledge, the proposed 1D-DCNN algorithm is firstly applied in the mixture gases recognition. Compared with the conventional pattern recognition algorithms including support vector machine (SVM) and artificial neural network (ANN), the proposed 1D-DCNN exhibits higher average accuracy (96.15%) based on extensive experimental results using 10-fold cross validation.

In this paper, we present a novel fast gas recognition algorithm based on convolutional recurrent neural network. Rectangular kernels and long short-term memory (LSTM) are adopted to calculate the gas-label distribution from relationship between sensors and relationship among time as well. Compared with the widely adopted algorithms, such as support vector machine (SVM), K nearest neighbor (KNN), random forest (RF), the proposed implementation features an ultra-short detection time of 4 seconds. Moreover, according to our extensive experimental results, a high accuracy of 98.28% is achieved, which greatly outperforms the abovementioned algorithms.

This paper presents a novel K times singular value decomposition (K-SVD) based denoising algorithm for the division-of-focal-plane (DoFP) polarization image sensors. In the proposed implementation, the input DoFP image can be expressed by the optimum sparse combination of the dictionary elements via K-SVD and orthogonal matching pursuit (OMP) algorithms. As a result, this implementation is capable of eliminating the Gaussian noise significantly and well-preserving the details and edges of the target DoFP image. Our extensive experimental results on various test images show that the proposed algorithm yields better visual quality and maintains a lower PSNR value while compared with a wide range of previous implementations.

Conventional resistive sensor readout ICs (ROICs) suffer limitations, either because of their high resistive sensor-biasing current consumption or low resolution for jitter-induced noise. To overcome these limitations, a 13-bit resistive sensor ROIC is presented in this work. In the proposed ROIC, which relies on zero-crossing-based circuits (ZCBCs), the current source is used not only to settle the output of the ΔΣ integrator, but also to bias the resistive sensor, and thus avoid implementing a dedicated power-hungry bias current branch. In addition, the oversampled ΔΣ modulation technique guarantees high resolution for the proposed ROIC. A prototype was fabricated using 0.18 μm CMOS technology. The measurement results show the ROIC achieved a 13-bit root-mean-square noise equivalent resolution with an 8 kS/s conversion speed and a 480 fJ/conversion-step figure of merit. Moreover, it has good nonlinearity of less than 250 ppm and a power supply sensitivity of 1%/V.

Conventional capacitance-to-digital converters (CDCs) suffer limitations either on narrow capacitance range or low resolution for jitter-induced noise and high power consumption. In order to overcome these limitations, a 13-bit 1 pF-to-10 nF generic CDC is presented. In the proposed CDC with the oversampled ΔΣ modulation, the zero-crossing-based circuits (ZCBCs) are used to replace the operational transconductance amplifier to avoid feedback loop stability issues. However, the ZCBCs inevitably incur the non-idealities and thus a novel calibration scheme is presented for efficient non-ideality-error cancellation. A prototype fabricated using 0.18 μm CMOS technology is experimentally verified using a MEMS capacitive humidity sensor. The measurement results show the CDC achieves a 13-bit rms noise equivalent resolution with a 128 μs conversion time and a 230 fJ/conversion-step figure of merit. The calibration scheme enhances the linearity from 7 bits to 11.4 bits in the 1 pF-to-10 nF compatible capacitance range.

Alternative polyadenylation (APA) can lead to various mRNA isforms differing in their 3' UTR, which contributes to the dynamics of gene regulation, including stability, localization and translation of mRNA. However, clustering of genes using poly(A) site data has not been extensively studied. Here we constructed a two-layer model based on canonical correlation analysis (CCA) to explore the clustering of genes with the consideration of the distribution and abundance of APA sites. We adopted the widely used hierarchical clustering with Average-Linkage (Hc-A), Complete-Linkage (Hc-C), and Single-Linkage (Hc-S) to comprehensively compare the proposed method with four distance measures including Pearson correlation coefficient (PCC), Euclidean distance (EUC), Manhattan distance (MAN), and Maximum distance (MAX) based on three performance indexes. Results showed that the proposed method performed better than other four distance measures and could generate clusters with more biological meaning and underlying molecular mechanisms.

For multiplierless FIR filter design and implementation, optimization of the product accumulation block (PAB) in transposed direct form structure has been ignored by most of the research. In this work, the power consumption of the PABs of FIR filters is studied theoretically and experimentally. It is shown that the PAB contributes most of the total power consumption in multiplierless FIR filters. Two implementation methods are proposed to reduce the power consumption of the PAB. Simulation results show that the total power can be reduced significantly at the cost of some area overhead.

In this paper, we present a novel smoothness-based interpolation algorithm for the division of focal plane Polarimeters (DoFP). By calculating the divided blocks' variance that represents their local smoothness, the proposed algorithm well-balances between the traditional bilinear and bicubic interpolation algorithms. In addition, compared with the previously reported gradient-based interpolation algorithm which only indicates the image's directional change along 0°, 90°, 45° and 135°, the presented smoothness-based algorithm covers the variations along all the possible directions, leading to more accurate selection between the bilinear and bicubic algorithms. According to our extensive simulation results, the proposed implementation exhibits the lowest mean square error (MSE) for the test images among all the previously reported algorithms, including bilinear, bicubic and gradient-based interpolation algorithms.

This paper presents a frequency-mode capacitive sensor readout front-end operating in a wide temperature range. First, a biased varactor temperature compensation (BVTC) is proposed to compensate the aggregate temperature gradients from the sensor and the oscillator circuit, achieving a nullified temperature coefficient for the oscillation frequency. Second, a temperature coherent quantization (TCQ) approach is proposed to enhance the sensor' sensitivity and to provide a self-referenced clock for digitization, whereby the influence from the temperature effect of the clock is minimized through a hybrid down-conversion time-to-digital converter (DTDC). A prototype chip was fabricated using the 0.18-μm CMOS process and it was verified using a commercial 5.92-to-6.53-pF capacitive pressure sensor over a temperature range of −20°C to 120°C. Proven by experiments, the prototype presented as good as ±0.4% full-scale pressure error in the 140°C temperature range.

Support vector machine (SVM) now attracts increasing attention in gas classification due to its high performance towards small samples and nonlinearity problems of the dataset. Previously, the probable mismatch between the dataset and the training parameters determined by trial and error or grid search may hinder the exploration of the best result. In this paper, we propose a novel approach to estimate the most suitable training parameters, based on the inbreeding prevention of genetic algorithm (GA) by assigning the training model parameters of SVM as its chromosome. Treating the k-fold cross validation of SVM training as the objective function, our new method makes the population on the whole evolve towards the values that are more appropriate for the dataset. The inbreeding prevention mechanism (IPM) can protect the population from converging over-rapidly before reaching the optimum value. Compared with the standard SVM, the proposed method has greatly improved the prediction accuracy in both training data and testing data.

The level shifters are crucial primitives in the multi-supply voltage circuits and systems. In this paper, an energy-efficient level shifter is proposed to achieve the conversion from the subthreshold voltage to the above threshold voltage. It is a hybrid structure consisting of the Wilson current mirror and the cross-coupled level shifter. By addressing the voltage drop issue of the level shifter based on the Wilson current mirror, the leakage power is significantly reduced, with the advantage of wide input voltage range for the Wilson current mirror level shifter well-preserved. In addition, the multi-threshold CMOS (MTCMOS) technology is employed to provide more flexibility for our ultra-low power design. The reported simulation results using 65 nm CMOS process validate our proposed implementation and an ultra-low power consumption of 19.44 fJ per conversion from 0.2 V to 1.2 V at 1 MHz is achieved without the need of any intermediate power supply.

The classic digital signal processing based on shannon sampling theory uniformly samples signals without making use of their intrinsic properties. Researches have proved that level-crossing sampling scheme sampling the signals in a non-uniform way by an asynchronous analog-to-digital converter (AADC) may lead to lower power consumption. In this paper, a new finite impulse response (FIR) filtering technique aiming to process the non-uniform sampled data and the corresponding system architecture are proposed. The design example shows that the computation complexity of the proposed one is lower than that of the existing non-uniform FIR filtering technique.

In the design of multiplierless FIR filters, researchers have made every effort to reduce the number of adders when coefficients multipliers are realized using adder-and-shift network to decrease the overall chip area. However, with the advance of IC technology, area becomes a less important issue than the speed. In this paper, we propose a speed oriented optimization of linear phase FIR filters, where the length of critical path is used as the criteria in the discrete coefficient search. The length of critical path is measured as the number of cascaded full adders rather than the traditional adder depth. Compared to the area oriented algorithm, the proposed algorithm can generate the filters with much shorter critical path delay and meanwhile the area-delay product is also reduced. Gate level simulations of benchmark filters verify the above claim.

Deterministic tree search algorithm for the design of multiplierless linear phase finite impulse response filters are usually time consuming. More researches therefore focus on how to restrict the number of discrete values assigned to each coefficient during a tree search. This paper proposes a polynomial-time tree search algorithm where each coefficient is fixed to only one discrete value. Due to the short search time, multiple searches with floating passband gain become possible, and each search may produce a feasible discrete solution. In such a way, if the whole floating passband gain range is partitioned into many smaller ones, a large amount of feasible discrete solutions may be obtained. In the end, all these feasible solutions are synthesized using an multiple constant multiplication algorithm, and the one using the least number of adders is the final design results. To accelerate the search, a low hardware cost scheme which fixes some coefficient values to 0 prior to each search is proposed. With these techniques, design examples show that the proposed algorithm significantly outperforms the existing algorithms in terms of design time while the hardware cost is kept low.

Sparse patterns for digital filters have been suggested to reduce the computational cost. However, the minimization of the number of non-zero coefficients under required filter order and frequency domain constraints is difficult to be accomplished in polynomial time in many cases. In this paper, a two-stage design based on Genetic Algorithm (GA) is proposed to search for the optimal solution with a given specification. A preliminary optimization stage is introduced to enhance the efficiency of the proposed GA. The proposed algorithm is evaluated through two sets of examples, which generate better results than existing algorithms.

In this work, a novel genetic algorithm (GA) is proposed for the design of multiplierless finite impulse response (FIR) filters with high filter order and wide coefficient wordlength. GA mimics the nature evolution to optimize complicated problems and in theory optimum solutions can be obtained with infinite computation time. However, in practical filter design problem, when the filter specification is stringent, requiring high filter order and wide coefficient wordlength, GA often fails to find feasible solutions, because the discrete search space thus constructed is huge and majority of the solution candidates therein can not meet the specification. In the proposed GA, the discrete search space is partitioned into smaller ones. Each of the small space is constructed surrounding an optimum continuous solution with a floating passband gain. This increases the chances for the GA to find feasible solutions, but not sacrificing the coverage of the search space. In addition, the search in the multiple spaces can run in parallel, and thus the computation time for the design of filters under consideration reduces significantly. Design examples show that the proposed GA outperforms existing algorithms dealing with the similar problems.

In this work, a novel algorithm is proposed for the design of low power linear phase finite impulse response (FIR) filters with discrete coefficients. The proposed algorithm finds an adder and shift network using a small logic depth to realize the coefficient multipliers. This is achieved by constraining the synthesis of each coefficient on its minimum adder depth. The new synthesis mechanism tends to use a few more adders to successfully realize the discrete filter coefficients than the current best algorithms, but with lower average logic depth, resulting in a low power implementation. Designs of benchmark filters and simulation results verified the above claims.

The switchings in the multiple constant multiplier (MCM) block of FIR filters are heavily affected by the spatial correlation of inputs at each computation unit—adders in this case. In this work, a new switching activity model for the MCM blocks is developed by taking the spatial correlation of inputs of each adder into consideration. Based on this model, a power measurement is proposed to estimate the power consumption of the MCM blocks of FIR filters. The dynamic power simulation of several benchmark filters shows that the accuracy of our model outperforms that of existing models such as Glitch Path Count, Glitch Path Score and Power Index.